Body mass index and associated factors among refugees living in North Rhine-Westphalia, Germany: a cross-sectional study | BMC Nutrition

Study settings and participants

The FHS is a cross-sectional study administered by the Forschungskolleg “FlueGe” at Bielefeld University [28]. The FHS aimed to provide health data of refugees from the main countries of origin that contributed to the European refugee crisis in 2015 and 2016 in the region of East Westphalia-Lippe in North Rhine-Westphalia, Germany. The data were collected from February to November 2018. Personal interviews and physical examinations were carried out by trained interviewers. The questionnaire was translated into the following five languages by Kantar Public, a consulting and market research institute: Arabic, Farsi, Kurmanji, English, and German. Participants were recruited from shared and private accommodations in eight different locations in East Westphalia-Lippe. Municipal cooperation partners and social workers helped in the access to potential participants. The FHS included all individuals who were willing to participate (convenience sampling) and signed informed consent. Participants were not eligible for the study if they were younger than 18 years of age, could not speak Arabic, Kurmanji, Farsi, English, or German, or if they have been in Germany for more than 5 years. The concept and design of the FHS are described elsewhere [28].

Data collection and variables

Socio-demographic, lifestyle factors, and health information

After acquiring their consent, eligible participants were interviewed face-to-face by trained interviewers, in the participant’s native language if possible. Socio-demographic information included age, sex (male, female), nationality (Syrians, Afghans, Iraqis, Iranians, African countries, other), length of stay since arrival (< 12, 12–24, 25–36, > 36 months), marital status (unmarried, married, divorced, widowed), employment status (employed, unemployed) and education, which was inquired according to CASMIN [29]. At first, nine educational groups were identified, which resulted in a combination of school and vocational qualifications. For further analysis, the CASMIN index was used to categorize three groups: low (general elementary education and/or basic vocational qualification), medium (intermediate general qualification and/or intermediate vocational qualification), and high education (lower or higher tertiary education). Information on lifestyle factors included exercise (no exercise, less than 120 min, at least 120 min of exercise per week), smoking status (smoker, non-smoker), fruit and vegetable consumption (at least 5 servings, less than 5 servings on an average day), eating habits (more than 4 meals, 4 meals or less per day) as well as alcohol use categorized into the following four groups in accordance with the Epidemiological Survey of Substance Abuse [30]: dangerous consumption (the consumption of five or more alcoholic beverages per day (more than 60 grams (g) of pure alcohol for men and more than 40 g for women), risky consumption (the consumption of three or four alcoholic beverages per day (24-60 g of pure alcohol for men and 12-40 g for women) and low-risk consumption (the consumption of one or two alcoholic beverages per day (< 24 g of pure alcohol for men and < 12 g for women) on at least one day per month. Sobriety (abstinence) was defined as no consumption of alcoholic beverages. Furthermore, participants were asked to state whether a physician diagnosed them with some specific diseases. The following diseases were included in the analysis: depression, diabetes mellitus type II, dyslipidemia, and hypertension.

Following the collection of anthropometric measurements (see below), participants were asked whether their weight had changed (i) during their escape/journey to Germany and (ii) since their arrival in Germany, and if so, how many kilograms (kg) of weight they have gained or lost.

Anthropometric measurements

Data was collected on body height, weight, hip and waist circumference, with bare-foot participants in minimal clothing. Height was measured to the nearest centimeter (cm) using portable stadiometers. Weight was assessed to the nearest 0.1 kg using digital scales. Waist and hip circumferences were measured in cm to the nearest 0.1 cm, using inextensible anthropometric tape. Waist-to-hip ratio, abdominal obesity, and BMI were determined using definitions by the WHO [9, 31].

Body mass index

Body mass index was calculated by dividing a person’s weight in kg by his or her height, in meters squared (m2) [9]. The number generated from this equation was considered the individual’s BMI since arrival. Furthermore, BMI upon arrival was calculated by measured weight ± reported weight gained or lost since arrival (in kg) / measured height2 (in m) and BMI before the escape was measured by measured weight ± reported weight gained or lost since arrival ± weight gained or lost during escape (in kg) / measured height2 (in m). Finally, BMI was categorized into underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5–24.9 kg/m2), pre-obesity (overweight, BMI 25.0–29.9 kg/m2), or obesity (class I/II/III, BMI ≥30.0 kg/m2) in accordance with the WHO classification for adults [9].

Waist-to-hip ratio, abdominal obesity

The waist-to-hip ratio was calculated as waist circumference in cm divided by hip circumference in cm [31]. Abdominal obesity was defined as a waist-to-hip ratio above 0.90 for males and 0.85 for females [31].

Statistical analyses

Statistical analyses were performed using STATA version 15.1. In the first analysis step, the data were evaluated descriptively. Differences in BMI before the escape, upon arrival, and since arrival by sex and age group (18–24, 25–29, 30–39, and 40+ years) were analyzed using the Chi-squared test. Differences in BMI over time were analyzed using Wilcoxon signed-rank test. Linear regression using the backward elimination technique [32] with bias-corrected and accelerated (BCa) confidence interval (bootstrapping, based on 5000 samples) [33] was applied to assess the potential predictors of BMI since arrival. Cases presenting a missing value for at least one of the modeling variables were excluded from the analyses (listwise exclusion). In the regression model, the dependent variable was BMI since arrival and the independent variables were age, sex, nationality (as dummy variables (DV) to indicate the absence or presence of categorical effect [32]), marital status (DV), education (DV), employment status, length of stay since arrival (DV), exercise (DV), smoking status, alcohol use (DV), fruit and vegetable consumption, eating habits, health conditions (DV), waist-to-hip ratio, abdominal obesity, and BMI before the escape. All the independent variables were entered into the equation first (full model) and removed one at a time, starting with the variable with the highest p-value. We set a p-value threshold at 0.05 to determine when to stop removing variables from the model. Effects with p-values < 0.05 were considered statistically significant. Attention was paid to linear regression assumptions, interactions among variables of interest, and plausibility of the identified effects. The model finding procedure was repeated in a sub-sample with cases that reported weight gain since arrival.

Source link

Get in Touch


Please enter your comment!
Please enter your name here

For security, use of Google's reCAPTCHA service is required which is subject to the Google Privacy Policy and Terms of Use.

I agree to these terms.


Related Articles


Get in Touch


Latest Posts

Supportscreen tag